The Antiprotozoal Drug Pentamidine Ameliorates Experimentally Induced

Home , Pentamidine

Esposito et al. Journal of Neuroinflammation 2012, 9:277 JOURNAL OF http://www.jneuroinflammation.com/content/9/1/277 NEUROINFLAMMATION

RESEARCH Open Access The antiprotozoal drug pentamidine ameliorates experimentally induced acute colitis in mice Giuseppe Esposito1*, Elena Capoccia1, Giovanni Sarnelli2, Caterina Scuderi1, Carla Cirillo2,3, Rosario Cuomo2 and Luca Steardo1

Abstract Background: Intestinal inflammation is partly driven by enteroglial-derived S100B protein. The antiprotozoal drug pentamidine directly blocks S100B activity. We aimed to investigate the effect of pentamidine on intestinal inflammation using an animal model of dextran sodium sulphate (DSS)-induced acute colitis. Methods: Mice were divided into: control group, colitis group (4% DSS for four days) and two pentamidine-treated colitis groups (0.8 mg/kg and 4 mg/kg). Anti-inflammatory effect of pentamidine was assessed in colonic tissue by evaluating the disease activity index and the severity of histological changes. Colonic tissue were also used to evaluate cyclooxigenase-2, inducible nitric oxide synthase, S100B, glial fibrillary acidic protein, phosphorylated-p38 MAPkinase, p50, p65 protein expression, malondyaldheyde production, mieloperoxidase activity, and macrophage infiltration. Nitric oxide, prostaglandin E2, interleukin-1 beta, tumor necrosis factor alpha, and S100B levels were detected in plasma samples. Parallel measurements were performed in vitro on dissected mucosa and longitudinal muscle myenteric plexus (LMMP) preparations after challenge with LPS + DSS or exogenous S100B protein in the presence or absence of pentamidine. Results: Pentamidine treatment significantly ameliorated the severity of acute colitis in mice, as showed by macroscopic evaluation and histological/biochemical assays in colonic tissues and in plasma. Pentamidine effect on inflammatory mediators was almost completely abrogated in dissected mucosa but not in LMMP. Conclusions: Pentamidine exerts a marked anti-inflammatory effect in a mice model of acute colitis, likely targeting S100B activity. Pentamidine might be an innovative molecule to broaden pharmacological tools against colitis. Keywords: Pentamidine, Acute colitis, S100B protein, Enteric glia

Background and protective functions toward enteric neurons and are Although the etiology of ulcerative colitis (UC) remains fully implicated in the modulation of neuronal activities incompletely understood, severe and persistent mucosal [5]. Moreover, EGC have been repeatedly reported to trig- infiltration of macrophages and neutrophils in the large ger and support intestinal inflammation [6] and to function intestine represents a prominent feature [1]. Immune as a first line of defense against pathogens [7]. EGC may cells release cytokines, interleukins and proinflammatory proliferate and be activated in response to injury and in- signaling molecules [2-4]. In addition to the well-known flammation undergoing reactive gliosis (enterogliosis), a involvement of macrophages and neutrophils, other cell condition in which they release neurotrophins, growth fac- types have been recently reported to substantially con- tors and proinflammatory cytokines cross-talking with tribute to the onset and progression of the disease. En- other infiltrating immune cells [8]. Alterations in the teric glial cells (EGC) play a fundamental role in the homeostasis of the enteric nervous system are induced by maintenance of gut homeostasis since they have trophic reactive enterogliosis and are characterized by the massive overexpression and secretion of specific astroglial-derived signaling molecules such as S100B protein [9,10]. * Correspondence: [email protected] ++ ++ 1Department of Physiology and Pharmacology ‘Vittorio Erspamer’, University S100B is a diffusible, Ca /Zn -p53 binding protein SAPIENZA of Rome, P. le Aldo Moro 5, 00185, Rome, Italy playing a pivotal role during intestinal inflammation, Full list of author information is available at the end of the article

© 2012 Esposito et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 2 of 12 http://www.jneuroinflammation.com/content/9/1/277

since it orchestrates proinflammatory signals [11,12]. Disease activity index (DAI) Aberrant expression and release of S100B correlate with The DAI scale is based on the evaluation of different the inflammatory status of the gut. S100B accumulates parameters characterizing experimental colitis induction at the RAGE (receptor for advanced glycation end pro- and progression. Body weight, presence of gross blood in ducts) site only in micromolar concentrations [11,13-15] the feces and stool consistency were recorded daily and such interaction leads to mitogen-activated protein (from day 0 to 7) by an observer blinded to the treat- kinase (MAPK) phosphorylation and consequent nuclear ment. According to the criteria proposed by Cooper factor-kappaB (NF-κB) activation, which, in turn, promote et al. [19], the DAI was determined by scoring changes the transcription of different cytokines and inducible nitric in: weight loss (0 = none; 1 = 1 to 5%; 2 = 5 to 10%; 3 = oxide synthase (iNOS) protein [14]. Molecular targeting of 10 to 20%; 4 = >20%); stool consistency (0 = normal; 2 = S100B protein during intestinal inflammation might there- loose; 4 = diarrhea) and rectal bleeding (0 = normal; 2 = fore represent an innovative approach to treat UC. occult bleeding; 4 = gross bleeding). At the end of the Pentamidine isethionate, discovered to have antiproto- experiment, mice were sacrificed and colons and spleens zoal activity in 1938, and approved in the United States were isolated to measure the length and weight of colon for the treatment of Pneumocystis carinii pneumonia and spleen, respectively. and other protozoal diseases [16], appears to be an in- triguing candidate. In addition to its antiprotozoal activ- Preparation of cytosolic extracts and western blot ity, pentamidine has been reported to inhibit S100B analysis activity because of its ability to block the interaction at Removed colonic tissues were processed for western blot the Ca++/p53 site of the protein [17]. analysis. Briefly, after homogenization in ice-cold hypo- Based on this background, the present study was tonic lysis buffer, protein concentration was determined aimed at evaluating the beneficial effect of a daily ad- using Bio-Rad protein assay kit (Bio-Rad, Milan, Italy). ministration of pentamidine in an acute model of UC Analysis of cyclooxigenase (COX)-2, iNOS, TNF-α, S100B, induced by dextran sulphate sodium (DSS) administra- glial fibrillary acidic protein (GFAP), phosphorylated-p38 tion in drinking water in CD-1 mice. DSS-induced colitis (p-p38) MAPK, p50, p65 and β-actin protein expression is highly reproducible and is a well-known in vivo model was performed on total protein fractions of homogenates. of experimental colitis in rodents that reproduces many Equivalent amounts (50 μg) of each homogenate under- features of UC [18]. We tested the effect of pentamidine went electrophoresis through a polyacrilamide minigel. on (i) intensity of the symptoms (diarrhea, blood in the Proteins were then transferred onto nitrocellulose mem- feces, animal weight loss) through a disease activity index brane that were saturated by incubation with 10% nonfat (DAI) scale [19]; (ii) release of cytokines and proinflamma- dry milk in 1× PBS overnight at 4°C and then incubated tory signaling molecules present in mice plasma; (iii) post- with either mouse anti-S100B (1:200 v/v, Neo-Marker, mortem evaluation of macroscopic shortening of large Milan, Italy), mouse anti-iNOS (1:2000 v/v, BD Bios- intestine and spleen weight; (iv) global colonic inflamma- ciences, Milan, Italy), rabbit anti-COX-2 (1:250 v/v, BD tion by the evaluation of biochemical and histological Biosciences), GFAP (1:5000 v/v, Abcam, Cambridge, UK), changes of the tissue. mouse anti-p50, mouse anti-p65 (1:1000 v/v, Santa Cruz Biotechnology, Santa Cruz, CA, USA), or mouse anti-β- Methods actin (1:1,000 v/v, Santa Cruz Biotechnology) for 2 h at Animals and experimental design room temperature (RT). Membranes were then incubated Six-weeks-old male CD-1 mice (25 to 35 g; Harlan La- with anti-mouse or anti-rabbit immunoglobulins coupled boratories, Udine, Italy) were used for the experiments. to peroxidase (1:2000 v/v, Dako, Milan, Italy). Immune Animals were randomly divided into five groups (n = 10 complexes were revealed by using enhanced chemilumin- each): noncolitic control group; colitic group; colitic escence detection reagents (Amersham Biosciences, Milan, group receiving daily pentamidine 0.8 mg/kg; colitic group Italy) and exposed to Kodak X-Omat film (Eastman Kodak receiving daily pentamidine 4 mg/kg; noncolitic group Co., Rochester, NY, USA OK). Protein bands were then receiving daily pentamidine 4 mg/kg (as drug internal scanned and densitometrically analyzed with a GS-700 im- control). Colitis was induced by administrating DSS aging densitometer. (4% w/v, MW 36,000 to 50,000) in drinking water for six consecutive days (starting from day 1), as described in Figure 1A. Pentamidine was given intraperitoneally Preparation of blood samples starting at day 2 through day 6. At day 7, animals were Before being sacrificed, mice were deeply anesthetized sacrificed and tissues were removed to perform macro- and the blood was taken by cardiac puncture and col- scopic, histochemical and biochemical analyses as lected in 5% EDTA vials. To determine nitric oxide described below. (NO), prostaglandin E2 (PGE2), IL-1β,TNF-α, and S100B Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 3 of 12 http://www.jneuroinflammation.com/content/9/1/277

Figure 1 (A) Dextran sulphate sodium (DSS)-exposed (4%) mice were treated daily with 0.8 mg/kg or 4 mg/kg pentamidine given intraperitoneally. Effect of pentamidine on (B) DAI score, (C) colonic length and (D) spleen weight in DSS-treated mice. Results are expressed as mean ± SEM of n = 5 experiments. ***P <0.001 vs. vehicle (saline); °P <0.05; °°P <0.01 and °°°P <0.001 vs. DSS. DAI, disease activity index.

levels, plasma was isolated from the blood, immediately Plasma PGE2, TNF-α, S100B, and IL-1β quantification frozen, and stored at −80°C until the assays. Quantitative determination of PGE2,IL-1β,TNF-α, and S100B was carried out performing enzyme linked- immunosorbent assay (ELISA) on plasma samples Plasma NO quantification (PGE2,TNF-α,IL-1β kits were from Invitrogen, Milan, - NOwasmeasuredasnitrite(NO2) accumulation in plasma. Italy; S100B kit was from BioVendor, Heidelberg, A spectrophotometer assay based on the Griess reaction Germany) according to the manufacturer’s kit instructions. was used [20]. Briefly, Griess reagent (1% sulphanilamide, 0.1% naphthylethylenediamine in H3PO4) was added to an equalvolumeofplasmaandtheabsorbancewasmeasured Macrophage infiltration in the mucosa - at 550 nm. NO2 concentration (nM) was thus determined Samples for immunohistochemical assessment were iso- using a standard curve of NaNO2. lated from distal colon and fixed in 4% paraformaldehyde, Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 4 of 12 http://www.jneuroinflammation.com/content/9/1/277

embedded in paraffin, sectioned in 15 μmslicesand mixture was heated at 80°C for 20 min. After cooling, processed for immunohistochemistry. Slices were pre- MDA formation was recorded (absorbance 530 nm and treated for 20 min using heat-mediated antigen retrieval absorbance 550 nm) in a Perkin Elmer (Waltham, MA, with a sodium citrate buffer, incubated with MAC387 [21] USA) spectrofluorimeter and the results were presented (1 μg/ml, Abcam) for 15 min at RT, and detected using as ng MDA/mL. horseradish peroxidase (HRP)-conjugated compact poly- mer system. DAB was used as the chromogen. Slices were Mucosa and LMMP preparations then counterstained with hematoxylin, mounted with In order to further evaluate the putative site and the Eukitt and analyzed with a microscope (Nikon Eclipse 80i mechanism of action of pentamidine, additional experi- by Nikon Instruments Europe, Amstelveen, Netherlands). ments were set up as follows. Colonic segments (ap- Images were captured by a high-resolution digital camera proximately 1.5 cm long) were isolated from another set (Nikon Digital Sight DS-U1). of CD-1 mice not previously used for any in vivo treatment. Mice were euthanized by injection of pentobar- p53 Immunofluorescence bital sodium (100 mg/kg) and distal colon was removed Colon slices derived from both control and treated mice and cut longitudinally to expose the mucosa. Under ster- were blocked in 10% bovine serum albumin 0.1% ile conditions, the tissues were placed in Dulbecco’s Triton-PBS solution for 90 min at RT and subsequently modified Eagle's medium (supplemented with 5% fetal stained for 1 h at RT with anti-p53 antibody (1:1250, bovine serum, 2 mM glutamine, 100 U/mL penicillin, Santa Cruz Biotechnology) and anti-MAC387 (1:200, 100 μg/mL streptomycin, all from Invitrogen), pinned Abcam). Sections were then incubated for 1 h at RT in flat and the mucosa was carefully peeled off to obtain the dark with the proper secondary antibody: fluorescein the longitudinal muscle-myenteric plexus (LMMP) layer isothiocyanate-conjugated anti-rabbit (1:100, Abcam) or [24]. Depending upon the experimental plan, both mu- Texas Red-conjugated anti-mouse (1:64, Abcam), re- cosa and LMMP were stimulated for 24 h with exogen- spectively. Nuclei were stained with Hoechst (1:5000, ous lipopolysaccharide (LPS) (10 μg/ml) + DSS (1% w/v) Sigma-Aldrich, Milan, Italy). Slides were thus analyzed or exogenous S100B (5 μM, Sigma-Aldrich, Milan, Italy), with a microscope (Nikon Eclipse 80i by Nikon Instru- with or without the addition of pentamidine (0.5 to ments Europe), and images were captured at 10X and 5 μM) administered to the tissue 10 min prior to LPS + 20X magnification by a high-resolution digital camera DSS or S100B stimulus. (Nikon Digital Sight DS-U1). Statistical analysis Myeloperoxidase (MPO) assay Results were expressed as mean ± SEM of experiments. MPO, a marker of polymorphonuclear leukocyte accu- Statistical analysis was performed using parametric one- mulation and general inflammation occurring in colonic way analysis of variance (ANOVA) and multiple com- tissue, was determined as previously described [22]. parisons were performed by Bonferroni's post hoc test. After removal, colonic tissue was rinsed with cold saline, P values <0.05 were considered significant. opened and deprived from the mucosa using a glass slide. The resulting layer was then homogenized in a so- Results lution containing 0.5% hexadecyltrimethylammonium Pentamidine ameliorates DAI score, preserves colonic bromide dissolved in 10 mM potassium phosphate buf- length and reduces splenomegaly induced by DSS fer (pH 7.0) and centrifuged for 30 min at 20000 × g at Starting from day 4 after DSS administration, DAI score 37°C. An aliquot of the supernatant was mixed with a was significantly increased in DSS groups. As expected, solution of tetramethylbenzidine (1.6 mM) and 0.1 mM DSS caused a consistent increase in bloody diarrhea to- H2O2. The absorbance was then spectrophotometrically gether with loss of body weight, as compared to control measured at 650 nm. MPO activity was determined as group (Figure 1B). DSS also caused a significant shorten- the amount of enzyme degrading 1 mmol of peroxide ing of colon and a marked increase of spleen weight per minute at 37°C and was expressed in milliunits per (Figure 1C and D). Pentamidine treatment inhibited the 100 mg of wet tissue weight. rise in DAI score in a dose-dependent way, suggesting an overall improvement of intestinal symptoms asso- Lipid peroxidation assay ciated with colitis after DSS administration; such effect Malonyl dialdehyde (MDA) was measured by the thio- was accompanied by a reduction of bloody stools, diar- barbituric acid colorimetric assay [23] in colonic tissue. rhea frequency, and a rescue of body weight (Figure 1B). Briefly, 1 mL 10% (w/v) trichloroacetic acid was added Moreover, pentamidine was able to preserve colonic to 450 μl of tissue lysate. After centrifugation, 1.3 mL length and to prevent splenomegaly in DSS-treated mice 0.5% (w/v) thiobarbituric acid was added and the (Figure 1C and D). Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 5 of 12 http://www.jneuroinflammation.com/content/9/1/277

Pentamidine inhibits intestinal inflammation induced by in S100B expression was accompanied by a significant DSS increase in lipid peroxidation and phosphorylation of In Figure 2 it is shown that the administration of DSS p38 MAPK (Figure 4A and B). These effects were caused a marked increase of iNOS, COX-2, S100B and related to both p50 and p65 protein overexpression GFAP protein expression compared to control mice. Pent- and indicated the activation of NF-κB(Figure4B).In amidine treatment resulted in a dose-dependent attenu- our experimental conditions, pentamidine significantly ation of iNOS, COX-2 and GFAP but not of S100B and dose-dependently inhibited MDA accumulation, protein overexpression (Figure 2A and B). As expected, p38 MAPK phosphorylation and, consequently, NF-κB Griess reaction and ELISA demonstrated that the admin- activation. istration of DSS caused a significant increase in plasma - NO2 level, PGE2,IL-1β,TNF-α and S100B, compared to Pentamidine reduces DSS-induced macrophage control mice (Figure 3). Also in this case and according to infiltration and MPO activity western blot results, pentamidine treatment caused a Colonic mucosa was extensively infiltrated by macro- marked and dose-dependent attenuation of all the inflam- phages in DSS-treated mice compared to control mice matory mediators in plasma, except for S100B release that (Figure 5A and B). Such increase was accompanied as remained unaffected after pentamidine treatment. well by neutrophil infiltration, as showed by the increased MPO activity (Figure 5C). Pentamidine treatment caused Pentamidine inhibits S100B-induced lipid peroxidation, a significant and dose-dependent reduction in macro- p38 MAPK phosphorylation, and NF-κB activation induced phage infiltration together with significant and dose- by DSS dependent inhibition of MPO activity, indicating that Further characterizing the inhibitory effect of pentami- pentamidine is able to control neutrophil infiltration in dine, we found that in DSS-treated mice the increase colonic tissues (Figure 5A and B).

Figure 2 (A) Western blot analysis showing the effect of pentamidine on iNOS, COX-2, GFAP and S100B protein expression in colonic tissue of DSS-treated mice; (B) relative quantification of immunoreactive bands (arbitrary units). Results are expressed as mean ± SEM of n = 5 experiments performed in triplicate. ***P <0.001 vs. vehicle (saline); °°P <0.01 and °°°P <0.001 vs. DSS. COX2, cyclooxigenase-2; DSS, dextran sodium sulphate; GFAP, glial fibrillary acidic protein; iNOS, inducible nitric oxide synthase. Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 6 of 12 http://www.jneuroinflammation.com/content/9/1/277

- Figure 3 Effect of pentamidine on the release of inflammatory mediators (NO2, PGE2, IL-1β, TNF-α) and S100B in plasma of DSS-treated mice. Results are expressed as mean ± SEM of n = 5 experiments performed in triplicate. ***P <0.001 vs. vehicle (saline); °°P <0.01 and °°°P <0.001 vs. DSS. DSS, dextran sodium sulphate; IL-1β, interleukin-1 beta; NO, nitric oxide; PGE2, prostaglandin E2; TNF-α, tumor necrosis factor alpha.

Pentamidine induces p53 expression in infiltrating S100B-dependent effect of pentamidine macrophages An internal check for the expression of S100B protein Macrophage infiltration in the colonic tissues from was performed in both mucosal and LMMP preparations DSS-treated mice was confirmed by immunofluorence confirming that the protein was expressed exclusively in (Figure 6A). Pentamidine treatment significantly and dose- the LMMP layer (Figure 7A). Preparations were separ- dependently reduced macrophage infiltration (Figure 6A). ately challenged with DSS (1%) plus LPS (1 μg/ml) or Very interestingly, such decrease was accompanied by the S100B (5 μM) in the presence or absence of pentamidine reverse trend in p53 expression. In fact, p53 resulted nega- (0.5 to 5 μM). Stimulation with LPS + DSS caused a tive in the colon from control and DSS-treated mice, while marked proinflammatory response in the mucosa as well a strong immunopositivity was found in colonic tissues as in LMMP preparations, as demonstrated by the in- from pentamidine-treated mice (Figure 6). Moreover, p53 crease in nitrites level, PGE2 and TNF-α level compared expression was found in macrophages, as revealed from to respective nonstimulated preparations (Figure 7B and the co-expression with MAC387, a specific macrophage C). Very interestingly, pentamidine had no significant marker [21] (Figure 6). beneficial effect on the mucosa while it significantly and Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 7 of 12 http://www.jneuroinflammation.com/content/9/1/277

Figure 4 (A) Effect of pentamidine on MDA production in colonic tissue of DSS-treated mice; (B) immunoblots showing the effect of pentamidine on p-p38 MAPK, p50 and p65 protein expression in colonic tissue of DSS-treated mice; (C) relative quantification of immunoreactive bands. Results are expressed as mean ± SEM of n = 5 experiments performed in triplicate. ***P <0.001 vs. vehicle (saline); °P <0.05, °°P <0.01 and °°°P <0.001 vs. DSS. DSS, dextran sodium sulphate; MDA, malondialdehyde; p-p38 MAPK, phosphorylated-p38 MAPkinase.

concentration-dependently reduced the release of nitrites, as monoclonal anti-TNF-α antibody (infliximab and ada- PGE2 and TNF-α in LMMP preparations (Figure 7C). Ex- limumab) have been recently introduced in the therapy ogenous application of S100B to the mucosa caused a of relapsing inflammatory bowel disease with encour- significant increase in nitrites, PGE2, and TNF-α level aging results in the maintenance of remission [28]. How- (Figure 7B). As confirmation for the S100B-dependent ever, the long-term safety of these drugs, the possibility effect, pretreatment with pentamidine significantly and to induce severe side effects [29-31] together with the concentration-dependently reduced S100B-induced in- high costs of the therapy for the patients warrant novel flammation in the mucosa (Figure 7C). and alternative pharmacological approaches. We demonstrated that pentamidine efficiently and Discussion dose-dependently improved colitis. It caused attenuation The results of the present study indicate that pentami- of the DAI together with preservation of colonic length dine is capable to profoundly and beneficially impact on and reduction of splenomegaly induced by DSS-colitis. an animal model of acute colitis. The model we used has Besides inducing such macroscopic beneficial effects, been demonstrated to resemble UC features [18]. Pre- treatment with pentamidine also resulted in microscopic vailing therapies for UC include chronic administration amelioration of intestinal inflammation as demonstrated of glucocorticosteroids and mesalamine [25,26]. Steroids by the reduction of MPO activity, a marker of tissue are effective in the short-term treatment of acute flares neutrophils activation, and of macrophage mucosal infil- of UC but they are not suitable as a maintenance ther- tration. Very interestingly, we found that the treatment apy due to a variety of systemic adverse reactions [27]. with pentamidine induced p53 expression on infiltrating Sulfasalazine and its derivative 5-aminosalicylic acid are macrophages. We know that S100B inhibits p53 activity effective only in mild-to-moderate acute phase of the and that pentamidine inhibits S100B overexpression. disease and in preventing relapse. Biological drugs such Thus, we may speculate that pentamidine, through the Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 8 of 12 http://www.jneuroinflammation.com/content/9/1/277

Figure 5 (A) Effect of pentamidine on macrophage infiltration in colonic tissue of DSS-treated mice: pentamidine dose-dependently reduced macrophages infiltration. Arrows indicate MAC387 immunopositive macrophages infiltrating colon criptae; (B) Quantification of MAC387 immunopositive macrophages in colon criptae. Data are expressed as mean ± SEM of n = 3 experiments; (C) effect of pentamidine on MPO level in colonic tissue of DSS-treated mice. Results are expressed as mean ± SEM of n = 6 experiments. ***P <0.001 vs. vehicle (saline); °°P <0.01 and °°°P <0.001 vs. DSS. Original magnification 100X. DSS, dextran sodium sulphate; MPO, mieloperoxidase.

inhibition of S100B, induces p53 expression indirectly Given the well-described inhibitory effect exerted by driving macrophages apoptosis. This trend could also ex- pentamidine on S100B activity [17], we evaluated plain the anti-inflammatory effect exerted by pentamidine. whether it could negatively control enteric gliosis, thus As expected, DSS treatment caused a marked increase exerting a negative modulation of intestinal inflamma- in COX-2 and iNOS protein expression in the colon tion. The administration of DSS demonstrated an enor- accompanied by a significant release of PGE2 and NO in mous increase of both S100B and GFAP protein the plasma. Such increase was significantly and dose- expression. Pentamidine was not capable of negatively dependently reduced by pentamidine treatment. We pre- modulating S100B protein expression in tissues and its viously demonstrated that during intestinal inflammation, release in plasma, while it was able to induce a signifi- S100B rises up to micromolar concentrations, further cant decrease in GFAP protein expression. This data amplifying inflammatory responses [6,8,12]. In these con- could be explained by the fact that since pentamidine ditions, S100B induces lipid peroxidation and activates inhibits the activity but not the expression/release of p38 MAPK phosphorylation leading, in turn, to NF-κB S100B [17], this drug may prevent the downstream activation and proinflammatory cytokines release. Here effects due to the over-release of S100B during intestinal again, we report that pentamidine treatment resulted in a inflammation. In an autocrine manner, S100B induces significant inhibition of lipid peroxidation, p38MAPK acti- proliferation of glial cells and, thus, increase in GFAP vation and relative stabilization of NF-κB in the cytoplasm expression. We can thus speculate that pentamidine, by targeting S100B activity. These results further confirm by the inhibition of S100B activity, may control glial that pentamidine, by interfering at the p53 binding site on proliferation as highlighted by the decrease in GFAP S100B protein, drastically reduces inflammatory events expression. Alternatively, the ability of pentamidine to induced by S100B, leading to the overall improvement of significantly reduce the expression of GFAP may be intestinal inflammation. secondary to the reduced levels of inflammatory Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 9 of 12 http://www.jneuroinflammation.com/content/9/1/277

Figure 6 (A) Immunofluorescence analysis of colonic tissues showing the effect of pentamidine on macrophage infiltration induced by DSS: pentamidine dose-dependently inhibited macrophage infiltration (MAC387, red) and increased p53 expression (green) in colon criptae. Arrows indicate MAC387/p53 co-expression; (B) relative quantification (MAC387+/p53 + area/total area percentage) of macrophage infiltration and p53 expression in DSS-treated mice. Data are expressed as mean ± SEM of n = 3 experiments. Original magnification 200X. DSS, dextran sodium sulphate.

cytokines that have been previously demonstrated to organotypic cultures of isolated mucosa or LMMP pre- drive glial proliferation [32]. parations [24] and we challenged them with DSS plus Since S100B release has been reported to increase LPS in the presence or absence of pentamidine. Our macrophage activity causing proinflammatory responses results demonstrated that pentamidine was able to sig- [33], we wondered whether the anti-inflammatory effect nificantly and concentration-dependently inhibit LPS + - exerted by pentamidine on acute intestinal inflammation DSS-induced NO2, PGE2, IL-1β and TNF-α release in could be the result of the direct targeting of S100B in LMMP but not in mucosal preparations. This result sug- the glial network or the consequence of the indirect ef- gests that pentamidine needs S100B to exert its anti- fect on infiltrating immune cells in the mucosa. We used inflammatory properties. To confirm such hypothesis, Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 10 of 12 http://www.jneuroinflammation.com/content/9/1/277

Figure 7 (A) Western blot analysis showing S100B protein expression in mucosa (M) and in LMMP preparations. Effect of pentamidine - on inflammatory mediators (NO2, PGE2, IL-1β, TNF-α) in mucosa (B) and in LMMP (C) in the presence of different proinflammatory stimuli (LPS + DSS or exogenous S100B depending upon the experiments). Results are expressed as mean ± SEM of n = 5 experiments. ***P <0.001 vs. vehicle (saline); °P <0.05, °°P <0.01 and °°°P <0.001 vs. proinflammatory stimulus DSS, dextran sodium sulphate; IL-1β, interleukin-1 beta; LMMP, longitudinal-muscle myenteric plexus; LPS, lipopolysaccharide; NO, nitric oxide; TNF-α, tumor necrosis factor alpha. we challenged mucosa preparations with exogenous Conclusions S100B in the presence or absence of pentamidine. Inter- Pentamidine may be putatively proposed as a new drug estingly, we found that pentamidine significantly reversed able to control the acute phase of intestinal inflammation, S100B-induced release of inflammatory mediators, thus likely by acting on glial activation and thus inhibiting the demonstrating that anti-inflammatory properties of pent- deleterious cascade induced by S100B protein. Because of amidine are dependent on S100B. its well-known pharmacological and toxicological profile, Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 11 of 12 http://www.jneuroinflammation.com/content/9/1/277

at present, pentamidine might be regarded as a potential, 10. von Boyen GB, Schulte N, Pflüger C, Spaniol U, Hartmann C, Steinkamp M: innovative, manageable, and low-cost tool against colitis. Distribution of enteric glia and GDNF during gut inflammation. BMC Gastroenterol 2011, 11:3. 11. Cirillo C, Sarnelli G, Turco F, Mango A, Grosso M, Aprea G, Masone S, Cuomo Abbreviations R: Proinflammatory stimuli activates human-derived enteroglial cells and COX2: cyclooxigenase-2; DAI: disease activity index; DSS: dextran sodium induces autocrine nitric oxide production. Neurogastroenterol Motil 2011, sulphate; EGC: enteric glial cells; GFAP: glial fibrillary acidic protein; IL- 23:e372–e382. β 1 : interleukin-1 beta; iNOS: inducible nitric oxide synthase; 12. Esposito G, Cirillo C, Sarnelli G, De Filippis D, D'Armiento FP, Rocco A, LMMP: longitudinal muscle-myenteric plexus; LPS: lipopolysaccharide; Nardone G, Petruzzelli R, Grosso M, Izzo P, Iuvone T, Cuomo R: Enteric glial- κ MDA: malonyldialdehyde; MPO: mieloperoxidase; NF- B: nuclear factor- derived S100B protein stimulates nitric oxide production in celiac kappaB; NO: nitric oxide; PGE2: prostaglandin E2; p-p38 disease. Gastroenterology 2007, 133:918–925. MAPK: phosphorylated-p38 MAPkinase; RAGE: receptor for advanced 13. Adami C, Bianchi R, Pula G, Donato R: S100B-stimulated NO production by α glycation end products; TNF- : tumor necrosis factor alpha; UC: ulcerative BV-2 microglia is independent of RAGE transducing activity but colitis. dependent on RAGE extracellular domain. Biochim Biophys Acta 2004, 1742:169–177. Competing interests 14. Hofmann MA, Drury S, Fu C, Qu W, Taguchi A, Lu Y, Avila C, Kambham N, All authors declare that they have no competing interest. Bierhaus A, Nawroth P, Neurath MF, Slattery T, Beach D, McClary J, Nagashima M, Morser J, Stern D, Schmidt AM: RAGE mediates a novel proinflammatory axis: a central cell surface receptor for S100/calgranulin ’ Authors contribution polypeptides. Cell 1999, 97:889–901. GE and EC conceived the study; GE designed experiments; EC performed 15. Schmidt AM, Yan SD, Yan SF, Stern DM: The multiligand receptor RAGE as experiments; GS and CC analyzed the data and contributed to the critical a progression factor amplifying immune and inflammatory responses. revision of the research plan; CS collected the data. GE, LS and RC equally J Clin Invest 2001, 108:949–955. contributed to the design of the study and writing and final approval of the 16. Pearson RD, Hewlett EL: Pentamidine for the treatment of Pneumocystis manuscript. All authors read and approved the final manuscript. carinii pneumonia and other protozoal diseases. Ann Intern Med 1985, 103:782–786. Acknowledgements 17. Charpentier TH, Wilder PT, Liriano MA, Varney KM, Pozharski E, MacKerell The authors thank Dr. Daniele De Filippis and Prof. Teresa Iuvone for their AD Jr, Coop A, Toth EA, Weber DJ: Divalent metal ion complexes of precious help in editing the manuscript. S100B in the absence and presence of pentamidine. J Mol Biol 2008, 382:56–73. Author details 18. Park MY, Ji GE, Sung MK: Dietary kaempferol suppresses inflammation of 1Department of Physiology and Pharmacology ‘Vittorio Erspamer’, University dextran sulfate sodium-induced colitis in mice. Dig Dis Sci 2012, SAPIENZA of Rome, P. le Aldo Moro 5, 00185, Rome, Italy. 2Department of 57:355–363. Clinical and Experimental Medicine, University of Naples FEDERICO II, Via S. 19. Cooper HS, Murthy SN, Shah RS, Sedergran DJ: Clinicopathologic study of Pansini 5, 80131, Naples, Italy. 3Current address: Laboratory for Enteric dextran sulfate sodium experimental murine colitis. Lab Invest 1993, NeuroScience (LENS), TARGID, KU Leuven, Herestraat 49, 3000, Leuven, 69:238–249. Belgium. 20. Di Rosa M, Radomski M, Carnuccio R, Moncada S: Glucocorticoids inhibit the induction of nitric oxide synthase in macrophages. Biochem Biophys Received: 26 September 2012 Accepted: 19 November 2012 Res Commun 1990, 172:1246–1252. Published: 23 December 2012 21. Thoree V, Skepper J, Deere H, Pele LC, Thompson RP, Powell JJ: Phenotype of exogenous microparticle-containing pigment cells of the human References Peyer's patch in inflamed and normal ileum. Inflamm Res 2008, – 1. Monteleone G, Pallone F, MacDonald TT: Emerging immunological 57:374 378. targets in inflammatory bowel disease. Curr Opin Pharmacol 2011, 22. Mullane KM, Kraemer R, Smith B: Myeloperoxidase activity as a 11:640–645. quantitative assessment of neutrophil infiltration into ischemic – 2. Helper DJ, Rex DK: Inflammatory bowel disease. Endoscopy 2001, myocardium. J Pharmacol Methods 1985, 14:157 167. 33:140–146. 23. Mihara M, Uchiyama M: Determination of malonaldehyde precursor in – 3. Reinders CI, Herulf M, Ljung T, Hollenberg J, Weitzberg E, Lundberg JO, tissues by thiobarbituric acid test. Anal Biochem 1978, 86:271 278. Hellström PM: Rectal mucosal nitric oxide in differentiation of 24. Kindt S, Vanden Berghe P, Boesmans W, Roosen L, Tack J: Prolonged IL- inflammatory bowel disease and irritable bowel syndrome. 1beta exposure alters neurotransmitter and electrically induced Ca(2+) Clin Gastroenterol Hepatol 2005, 3:777–783. responses in the myenteric plexus. Neurogastroenterol Motil 2010, 4. Hendel J, Nielsen OH: Expression of cyclooxygenase-2 mRNA in active 22:321–e85. inflammatory bowel disease. Am J Gastroenterol 1997, 25. Ford AC, Bernstein CN, Khan KJ, Abreu MT, Marshall JK, Talley NJ, Moayyedi 92:1170–1173. P: Glucocorticosteroid therapy in inflammatory bowel disease: systematic 5. Ruhl A, Nasser Y, Sharkey KA: Enteric glia. Neurogastroenterol Motil 2004, review and meta-analysis. Am J Gastroenterol 2011, 106:590–599. 16:44–49. 26. Ford AC, Achkar JP, Khan KJ, Moayyedi P: Efficacy of 5-aminosalicylates in 6. Cirillo C, Sarnelli G, Esposito G, Turco F, Steardo L, Cuomo R: S100B protein ulcerative colitis: systematic review and meta-analysis. Am J Gastroenterol in the gut: the evidence for enteroglial-sustained intestinal 2011, 106:601–616. inflammation. World J Gastroenterol 2011, 17:1261–1266. 27. Cottone M, Renna S, Modesto I, Orlando A: Is 5-ASA still the treatment of 7. Collins SM, Bercik P: The relationship between intestinal microbiota and choice for ulcerative colitis? Curr Drug Targets 2011, 12:1396–1405. the central nervous system in normal gastrointestinal function and 28. Rutgeerts P, Vermeire S, Van Assche G: Biological therapies for disease. Gastroenterology 2009, 136:2003–2014. inflammatory bowel diseases. Gastroenterology 2009, 136:1182–1197. 8. De Filippis D, Esposito G, Cirillo C, Cipriano M, De Winter BY, Scuderi C, 29. Fok KC, Ng WW, Henderson CJ, Connor SJ: Cutaneous sarcoidosis in a patient Sarnelli G, Cuomo R, Steardo L, De Man JG, Iuvone T: Cannabidiol reduces with ulcerative colitis on infliximab. JCrohnsColitis2012, intestinal inflammation through the control of neuroimmune axis. 6:708–712. PLoS One 2011, 6:e28159. 30. Kinnunen U, Färkkilä M, Mäkisalo H: A case report: Ulcerative colitis, 9. Cirillo C, Sarnelli G, Esposito G, Grosso M, Petruzzelli R, Izzo P, Calì G, treatment with an antibody against tumor necrosis factor (infliximab), D'Armiento FP, Rocco A, Nardone G, Iuvone T, Steardo L, Cuomo R: and subsequent liver necrosis. J Crohns Colitis 2012, 6:724–727. Increased mucosal nitric oxide production in ulcerative colitis is 31. Huang X, Lv B, Jin HF, Zhang S: A meta-analysis of the therapeutic effects mediated in part by the enteroglial-derived S100B protein. of tumor necrosis factor-α blockers on ulcerative colitis. Eur J Clin Neurogastroenterol Motil 2009, 21:1209–e112. Pharmacol 2011, 67:759–766. Esposito et al. Journal of Neuroinflammation 2012, 9:277 Page 12 of 12 http://www.jneuroinflammation.com/content/9/1/277

32. von Boyen GB, Steinkamp M, Reinshagen M, Schäfer KH, Adler G, Kirsch J: Proinflammatory cytokines increase glial fibrillary acidic protein expression in enteric glia. Gut 2004, 53:222–228. 33. Esposito G, De Filippis D, Cirillo C, Sarnelli G, Cuomo R, Iuvone T: The astroglial-derived S100beta protein stimulates the expression of nitric oxide synthase in rodent macrophages through p38 MAP kinase activation. Life Sci 2006, 78:2707–2715.

doi:10.1186/1742-2094-9-277 Cite this article as: Esposito et al.: The antiprotozoal drug pentamidine ameliorates experimentally induced acute colitis in mice. Journal of Neuroinflammation 2012 9:277.

Submit your next manuscript to BioMed Central and take full advantage of:

• Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit